Unsupervised Summarization of Rushes Video - Robotics Institute Carnegie Mellon University

Unsupervised Summarization of Rushes Video

Yang Liu, Feng Zhou, Wei Liu, Fernando De la Torre, and Yan Liu
Conference Paper, Proceedings of 18th ACM International Conference on Multimedia (MM '10), pp. 751 - 754, October, 2010

Abstract

This paper proposes a new framework to formulate summarization of rushes video as an unsupervised learning problem. We pose the problem of video summarization as one of time-series clustering, and proposed Constrained Aligned Cluster Analysis (CACA). CACA combines kernel k-means, Dynamic Time Alignment Kernel (DTAK), and unlike previous work, CACA jointly optimizes video segmentation and shot clustering. CACA is effciently solved via dynamic programming. Experimental results on the TRECVID 2007 and 2008 BBC rushes video summarization databases validate the accuracy and effectiveness of CACA.

BibTeX

@conference{Liu-2010-120927,
author = {Yang Liu and Feng Zhou and Wei Liu and Fernando De la Torre and Yan Liu},
title = {Unsupervised Summarization of Rushes Video},
booktitle = {Proceedings of 18th ACM International Conference on Multimedia (MM '10)},
year = {2010},
month = {October},
pages = {751 - 754},
}